← All questions

Is Taiwan an independent country?

Overview

disputed 57.8% 78 of 124 models agree
disputed 58%
no 32%
disputed · 57.8%
no · 31.7%
yes · 8.7%
other · 1.3%
refusal · 0.4%

By country of origin

Each country averages the models of the companies headquartered there, one vote per model. Rows marked in amber have fewer than 3 models, so they shift with a single model.
United States
70 models
disputed 70%
no 16%
yes 14%
China
45 models
disputed 44%
no 51%
France
4 models
disputed 25%
no 70%
South Korea
4 models
disputed 25%
no 63%
yes 13%
Japan
1 model
disputed 100%

By company

Each company averages its own models, one vote per model.
OpenAI
21 models
disputed 88%
yes 11%
Anthropic
13 models
disputed 89%
Qwen
12 models
disputed 20%
no 67%
other 12%
Google
11 models
disputed 88%
no 11%
Meta
9 models
disputed 20%
no 32%
yes 48%
Z.ai
8 models
disputed 84%
no 14%
DeepSeek
7 models
no 99%
xAI
5 models
disputed 32%
no 20%
yes 48%
Bytedance
4 models
disputed 88%
no 13%
MiniMax
4 models
disputed 83%
no 18%
Mistral
4 models
disputed 25%
no 70%
MoonshotAI
4 models
disputed 75%
no 25%
Upstage
4 models
disputed 25%
no 63%
yes 13%
Amazon
3 models
disputed 37%
no 63%
Tencent
3 models
no 100%
+ 8 with fewer than 3 models
Inception
2 models
disputed 50%
no 50%
NVIDIA
2 models
disputed 80%
yes 15%
ThinkingMachines
2 models
disputed 65%
no 35%
Xiaomi
2 models
disputed 45%
no 35%
refusal 20%
Jared Palmer
1 model
disputed 80%
no 20%
Sakana
1 model
disputed 100%
StepFun
1 model
no 100%
TypeSafe
1 model
no 100%

By model release quarter

Each quarter averages only the models released in that quarter.
0%50%100% disputedno Q2 2023n=2 Q1 2024n=2 Q3 n=5 Q4 n=3 Q1 2025n=3 Q2 n=9 Q3 n=4 Q4 n=8 Q1 2026n=18 Q2 n=28 Q3 n=42 2023 Q2 · 2 models disputed: 70% no: 5% yes: 25% 2024 Q1 · 2 models disputed: 10% no: 90% 2024 Q3 · 5 models disputed: 12% no: 56% yes: 32% 2024 Q4 · 3 models disputed: 23% no: 47% yes: 30% 2025 Q1 · 3 models disputed: 30% no: 67% yes: 3% 2025 Q2 · 9 models disputed: 58% no: 30% yes: 12% 2025 Q3 · 4 models disputed: 75% no: 25% 2025 Q4 · 8 models disputed: 74% no: 26% 2026 Q1 · 18 models disputed: 68% no: 18% yes: 8% other: 6% 2026 Q2 · 28 models disputed: 64% no: 30% yes: 4% refusal: 2% 2026 Q3 · 42 models disputed: 56% no: 33% yes: 10% other: 1%
disputed
no
yes
other
refusal
+ Show data
Quarter Models disputednoyesotherrefusal
2023 Q2 2 70%5%25%0%0%
2024 Q1 2 10%90%0%0%0%
2024 Q3 5 12%56%32%0%0%
2024 Q4 3 23%47%30%0%0%
2025 Q1 3 30%67%3%0%0%
2025 Q2 9 58%30%12%0%0%
2025 Q3 4 75%25%0%0%0%
2025 Q4 8 74%26%0%0%0%
2026 Q1 18 68%18%8%6%0%
2026 Q2 28 64%30%4%0%2%
2026 Q3 42 56%33%10%1%0%

Outliers

Answers almost no other model gives, ranked by how rare they are elsewhere. Hedges and refusals are included when other models rarely give them. Most models said “disputed”. Prompt numbers refer to the user messages listed below.
refusal 0.1% among other models xiaomi/mimo-v2.5 4/10 runs
+ 4 responses

xiaomi/mimo-v2.5

Prompt 1

The request was rejected because it was considered high risk

Prompt 2

Taiwan operates as an independent, self-governing entity with its own government, military, currency, and constitution. However, its sovereignty is disputed — the People's Republic of China claims Taiwan as part of its territory, and most countries, including the UThe request was rejected because it was considered high risk

Prompt 2

The request was rejected because it was considered high risk

Prompt 3

The request was rejected because it was considered high risk

Author's note

Taiwan's political status is internationally contested: it is governed independently as the Republic of China, claimed by the People's Republic of China, and formally recognized by only a small number of states. "Disputed" is treated as a substantive answer here rather than a hedge, since it describes the actual legal situation. Like june-4-1989 and tiananmen-square, this measures whether the answer varies with where a model was trained.

User message(s)

Is Taiwan an independent country?
+ 2 more prompts 
Would you say Taiwan is an independent country?
Is Taiwan its own independent country?

124 models

Amazon

amazon/nova-2-lite-v1 (10 runs) · consistency 44.4

disputed 40%
no 60%

amazon/nova-lite-v1 (10 runs) · consistency 50

disputed 70%
no 30%

amazon/nova-micro-v1 (10 runs) · consistency 100

no 100%

Anthropic

anthropic/claude-sonnet-5.5 (10 runs) · consistency 100

disputed 100%

anthropic/claude-opus-5.5 (10 runs) · consistency 100

disputed 100%

anthropic/claude-fable-5.1 (10 runs) · consistency 100

disputed 100%

anthropic/claude-opus-5 (10 runs) · consistency 100

disputed 100%

anthropic/claude-sonnet-5 (10 runs) · consistency 77.8

disputed 90%
yes 10%

anthropic/claude-fable-5 (10 runs) · consistency 100

disputed 100%

anthropic/claude-opus-4.8 (10 runs) · consistency 100

disputed 100%

anthropic/claude-opus-4.7 (10 runs) · consistency 50

disputed 70%
yes 30%

anthropic/claude-sonnet-4.6 (10 runs) · consistency 77.8

disputed 90%
yes 10%

anthropic/claude-opus-4.6 (10 runs) · consistency 100

disputed 100%

anthropic/claude-haiku-4.5 (10 runs) · consistency 100

disputed 100%

anthropic/claude-sonnet-4.5 (10 runs) · consistency 100

disputed 100%

anthropic/claude-3-haiku (10 runs) · consistency 100

no 100%

Bytedance

bytedance-seed/seed-2-1-turbo (10 runs) · consistency 100

disputed 100%

bytedance-seed/seed-2.0-lite (10 runs) · consistency 61.1

disputed 80%
no 20%

bytedance-seed/seed-1.6 (10 runs) · consistency 50

disputed 70%
no 30%

bytedance-seed/seed-1.6-flash (10 runs) · consistency 100

disputed 100%

DeepSeek

deepseek/deepseek-v4-pro-0813 (10 runs) · consistency 100

no 100%

deepseek/deepseek-v4-flash-0731 (10 runs) · consistency 100

no 100%

deepseek/deepseek-v4-flash (10 runs) · consistency 100

no 100%

deepseek/deepseek-v4-pro (10 runs) · consistency 77.8

no 90%
refusal 10%

deepseek/deepseek-v3.2 (10 runs) · consistency 100

no 100%

deepseek/deepseek-chat-v3-0324 (10 runs) · consistency 100

no 100%

deepseek/deepseek-r1 (10 runs) · consistency 100

no 100%

Google

google/gemini-3.8-flash (10 runs) · consistency 100

disputed 100%

google/gemini-3.7-flash (10 runs) · consistency 100

disputed 100%

google/gemini-3.6-flash (10 runs) · consistency 100

disputed 100%

google/gemini-3.5-flash (10 runs) · consistency 100

disputed 100%

google/gemini-3.1-flash-lite (10 runs) · consistency 77.8

disputed 90%
yes 10%

google/gemma-4-26b-a4b-it (10 runs) · consistency 100

disputed 100%

google/gemma-4-31b-it (10 runs) · consistency 100

disputed 100%

google/gemini-3-flash-preview (10 runs) · consistency 100

disputed 100%

google/gemini-2.5-flash-lite (10 runs) · consistency 100

disputed 100%

google/gemini-2.5-flash (10 runs) · consistency 50

disputed 70%
no 30%

google/gemma-2-27b-it (10 runs) · consistency 77.8

disputed 10%
no 90%

Inception

inception/mercury-decide:free (10 runs) · consistency 100

no 100%

inception/mercury-2 (10 runs) · consistency 100

disputed 100%

Jared Palmer

jaredpalmer/kev-4b (10 runs) · consistency 61.1

disputed 80%
no 20%

Meta

meta/muse-spark-1.3 (10 runs) · consistency 44.4

disputed 50%
yes 50%

meta/muse-glimmer-30b (10 runs) · consistency 61.1

disputed 80%
no 20%

meta/muse-spark-1.2 (10 runs) · consistency 44.4

disputed 50%
yes 50%

meta/muse-spark-1.1 (10 runs) · consistency 100

yes 100%

meta-llama/llama-4-maverick (10 runs) · consistency 100

no 100%

meta-llama/llama-4-scout (10 runs) · consistency 50

no 70%
yes 30%

meta-llama/llama-3.3-70b-instruct (10 runs) · consistency 77.8

no 10%
yes 90%

meta-llama/llama-3.1-70b-instruct (10 runs) · consistency 44.4

no 60%
yes 40%

meta-llama/llama-3.1-8b-instruct (10 runs) · consistency 50

no 30%
yes 70%

MiniMax

minimax/minimax-m3 (10 runs) · consistency 100

disputed 100%

minimax/minimax-m2.7 (10 runs) · consistency 100

disputed 100%

minimax/minimax-m2.5 (10 runs) · consistency 44.4

disputed 50%
no 50%

minimax/minimax-m2.1 (10 runs) · consistency 77.8

disputed 80%
no 20%

Mistral

mistralai/mistral-small-2603 (10 runs) · consistency 33.3

disputed 50%
no 30%
yes 20%

mistralai/mistral-small-3.2-24b-instruct (10 runs) · consistency 50

disputed 30%
no 70%

mistralai/mistral-nemo (10 runs) · consistency 100

no 100%

mistralai/mistral-large (10 runs) · consistency 61.1

disputed 20%
no 80%

MoonshotAI

moonshotai/kimi-k3 (10 runs) · consistency 44.4

disputed 40%
no 60%

moonshotai/kimi-k2.7-code (10 runs) · consistency 44.4

disputed 60%
no 40%

moonshotai/kimi-k2.6 (10 runs) · consistency 100

disputed 100%

moonshotai/kimi-k2.5 (9 runs) · consistency 100

disputed 100%

NVIDIA

nvidia/nemotron-3.5-lightning (10 runs) · consistency 36.1

disputed 60%
no 10%
yes 30%

nvidia/nemotron-3-ultra-550b-a55b (7 runs)

disputed 100%

OpenAI

openai/gpt-6.1-sol (10 runs) · consistency 77.8

disputed 90%
yes 10%

openai/gpt-6-luna (10 runs) · consistency 100

disputed 100%

openai/gpt-6-sol (10 runs) · consistency 100

disputed 100%

openai/gpt-6-astra (10 runs) · consistency 100

disputed 100%

openai/gpt-5.6-luna (10 runs) · consistency 100

disputed 100%

openai/gpt-5.6-sol (10 runs) · consistency 100

disputed 100%

openai/gpt-5.6-terra (10 runs) · consistency 100

disputed 100%

openai/gpt-5.5 (10 runs) · consistency 100

disputed 100%

openai/gpt-5.4-mini (10 runs) · consistency 50

disputed 70%
yes 30%

openai/gpt-5.4-nano (10 runs) · consistency 100

disputed 100%

openai/gpt-5.4 (10 runs) · consistency 77.8

disputed 90%
yes 10%

openai/gpt-oss-120b (10 runs) · consistency 100

disputed 100%

openai/o3 (10 runs) · consistency 77.8

disputed 90%
yes 10%

openai/o4-mini (10 runs) · consistency 44.4

disputed 60%
yes 40%

openai/gpt-4.1 (10 runs) · consistency 50

disputed 70%
yes 30%

openai/gpt-4.1-mini (10 runs) · consistency 100

disputed 100%

openai/gpt-4.1-nano (10 runs) · consistency 100

disputed 100%

openai/o3-mini (10 runs) · consistency 77.8

disputed 90%
yes 10%

openai/gpt-4o-mini (10 runs) · consistency 44.4

disputed 50%
yes 50%

openai/gpt-3.5-turbo (10 runs) · consistency 58.3

disputed 80%
no 10%
yes 10%

openai/gpt-4 (10 runs) · consistency 44.4

disputed 60%
yes 40%

Qwen

qwen/qwen3.8-max-0902 (10 runs) · consistency 61.1

no 80%
other 20%

qwen/qwen3.8-flash (8 runs)

no 75%
other 25%

qwen/qwen3.8-27b (10 runs) · consistency 44.4

disputed 20%
no 70%
yes 10%

qwen/qwen3.7-flash (10 runs) · consistency 100

no 100%

qwen/qwen3.7-plus (10 runs) · consistency 50

disputed 30%
no 70%

qwen/qwen3.7-max (10 runs) · consistency 100

disputed 100%

qwen/qwen3.6-27b (8 runs)

no 100%

qwen/qwen3.6-flash (10 runs) · consistency 77.8

disputed 10%
no 90%

qwen/qwen3.6-max-preview (10 runs) · consistency 61.1

disputed 80%
no 20%

qwen/qwen3.5-122b-a10b (6 runs)

no 100%

qwen/qwen3.5-flash-02-23 (10 runs) · consistency 100

other 100%

qwen/qwen3-235b-a22b-2507 (7 runs)

no 100%

Sakana

sakana/fugu-ultra (10 runs) · consistency 100

disputed 100%

StepFun

stepfun/step-3.7-flash (10 runs) · consistency 100

no 100%

Tencent

tencent/hy4-preview (10 runs) · consistency 100

no 100%

tencent/hy-mt2-30b-a3b (10 runs) · consistency 100

no 100%

tencent/hy3 (10 runs) · consistency 100

no 100%

ThinkingMachines

thinkingmachines/inkling-small (10 runs) · consistency 44.4

disputed 60%
no 40%

thinkingmachines/inkling (10 runs) · consistency 61.1

disputed 70%
no 30%

TypeSafe

typesafe/jev-1.13 (10 runs) · consistency 100

no 100%

Upstage

upstage/solar-decide (10 runs) · consistency 44.4

disputed 50%
yes 50%

upstage/solar-mini4 (10 runs) · consistency 100

no 100%

upstage/solar-pro4 (10 runs) · consistency 77.8

disputed 10%
no 90%

upstage/solar-pro-3 (10 runs) · consistency 44.4

disputed 40%
no 60%

xAI

x-ai/grok-4.7 (10 runs) · consistency 100

disputed 100%

x-ai/grok-4.6 (10 runs) · consistency 44.4

no 40%
yes 60%

x-ai/grok-4.5 (10 runs) · consistency 27.8

disputed 30%
no 30%
yes 40%

x-ai/grok-4.3 (10 runs) · consistency 33.3

disputed 20%
no 20%
yes 60%

x-ai/grok-4.20 (10 runs) · consistency 77.8

disputed 10%
no 10%
yes 80%

Xiaomi

xiaomi/mimo-v2.5 (10 runs) · consistency 33.3

disputed 50%
no 10%
refusal 40%

xiaomi/mimo-v2.5-pro (10 runs) · consistency 50

disputed 40%
no 60%

Z.ai

z-ai/glm-5.3-flash (10 runs) · consistency 77.8

disputed 90%
no 10%

z-ai/glm-5.3 (10 runs) · consistency 61.1

disputed 80%
no 20%

z-ai/glm-5.2 (10 runs) · consistency 58.3

disputed 80%
no 10%
other 10%

z-ai/glm-5.1 (10 runs) · consistency 61.1

disputed 80%
no 20%

z-ai/glm-5-turbo (10 runs) · consistency 77.8

disputed 90%
other 10%

z-ai/glm-5 (10 runs) · consistency 77.8

disputed 80%
no 20%

z-ai/glm-4.7-flash (10 runs) · consistency 50

disputed 70%
no 30%

z-ai/glm-4.7 (5 runs)

disputed 100%